scope-recon
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFE
Full Analysis
- [DATA_INGESTION_SURFACE]: The skill processes potentially untrusted data from local project files and external web sources. While this presents an indirect prompt injection surface, it is standard for this type of agent functionality.
- Ingestion points: Project documents (Step 1), external websites via WebSearch/WebFetch (Step 1), and the formatting guide at docs/output-kit.md (Step 3).
- Boundary markers: None present to distinguish untrusted content from system instructions.
- Capability inventory: The skill has access to Bash, Write, and web tools (WebSearch, WebFetch).
- Sanitization: No explicit validation or filtering of the processed content is specified.
Audit Metadata